PROJECT 3 ABSTRACT/SUMMARY The majority of pediatric cancers, and a large subset of adult malignancies, have relatively few mutated neo- epitopes due to low mutation burden, and few infiltrating lymphocytes, rendering immune checkpoint inhibition (ICI) strategies ineffective. Here, we hypothesize that many of the immunodominant antigens recognized on cold pediatric tumors will not be mutation-driven neoantigens, but rather primitive tissue-restricted fetal-differentiation antigens, shared between patients with similar cancers. We also hypothesize that tumors of similar origin in humans and mice share homologues of these immunodominant non-mutated protein antigens, enabling the biology and immune-oncology of effective immunotherapy in immunocompetent mouse models to direct the development of novel immunotherapeutic strategies. Project 3 builds upon our recent immunotherapeutic and immunoproteomic advances and is designed to create more effective immunotherapies for high-risk pediatric cancers. We propose to extend our in situ vaccine approach, presently consisting of local radiotherapy, intratumoral IL2 and tumor reactive mAb, together with ICI for Treg depletion, that eradicates cold tumors in immunocompetent mice and induces immunologic memory. This strategy will now be applied to novel immunocompetent murine models of neuroblastoma (NBL), medulloblastoma (MBL), and sarcomas (SARC) in Aim 1, providing critical preclinical data needed to design more effective immunotherapy regimens. In Aim 2, tumors from treated animals in Aim 1 will be studied by MHC-enriched mass spectrometry, and serum from the same mice subjected to antibody detection array methods, to identify the immunodominant targets responsible for the induced adaptive immunity. Aim 3 will extend our unpublished preliminary data presented herein showing that MHC-enriched mass spectrometry of human NBLs identifies not mutated neoantigens, but rather ?self? peptides from proteins with expression normally restricted to early development, to analyses of additional human NBL, as well as MBL and SARC samples, and demonstrate that these induce specific cytolytic T cell responses. Finally, in Aim 4, antibodies found in sera from pediatric cancer patients, and T cells from control donors and from pediatric patients, will be used to identify the immunodominant antigens found on the pediatric tumors studied in Aim 3. This Project is highly integrated with the overall Center. Proteogenomic data from Project 1 is required to determine the tumor specificity of the immunodominant antigens recognized on mouse and human pediatric tumors. With Project 2 we will determine if cellular therapies induce adaptive responses, and if this could be synergistically enhanced with our in situ vaccination approach developed in Aim 1. In addition, tumors responding and then escaping from in situ vaccine treatment will be studied for mechanisms of escape in Project 2. Taken together, the mechanisms and molecules identified here should be rapidly applied to the design and careful immunologic monitoring of next generation immunotherapy trials for childhood cancers. ! 1!